Package edu.msu.cme.rdp.classifier.train.validation

Examples of edu.msu.cme.rdp.classifier.train.validation.TreeFactory

31
32
33
34
35
36
37
38
39
40
41
    public MainMovingWindow(String taxFile, String inFile, String testFile , String outFile, String ecoliFile, int begin, int end, int  min_bootstrap_words) throws IOException{      
        windowFrames = this.getWindowFrame(ecoliFile);
              
        beginIndex = begin;
       
        factory = new TreeFactory(new FileReader(taxFile));           
         // create a tree
        createTree(factory, new File(inFile));                           
        outWriter = new BufferedWriter(new FileWriter(outFile));
        outWriter.write("taxon file: " + taxFile + "\n" + "train sequence file: " + inFile +"\n");       
       
View Full Code Here
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
    /** Creates a new Classification*/
    public LeaveOneOutTesterMain(String taxFile, String trainseqFile, String testFile, String outFile,
            int numGoodBases, int min_bootstrap_words, boolean hideTaxon) throws IOException {
        boolean useSeed = true;  // use seed for random word selection

        TreeFactory factory = new TreeFactory(new FileReader(taxFile));
        // create a tree
        createTree(factory, trainseqFile);

        BufferedWriter outWriter = new BufferedWriter(new FileWriter(outFile));
        LineageSequenceParser parser = new LineageSequenceParser(new File(testFile));
        LeaveOneOutTester tester = new LeaveOneOutTester(outWriter, numGoodBases);

        outWriter.write("taxon file: " + taxFile + "\n" + "train sequence file: " + trainseqFile + "\n");
        outWriter.write("word size: " + GoodWordIterator.getWordsize() + "\n");
        outWriter.write("minimum number of words for bootstrap: " + min_bootstrap_words + "\n");

        if (numGoodBases > 0) {    // do partial  

            outWriter.write("query sequence file: " + testFile + "\n"
                    + "classify partial sequence, number of good bases=" + numGoodBases + "\n");
        } else {
            outWriter.write("query sequence file: " + testFile + "\n"
                    + "classify full-length sequence \n");

        }
        outWriter.write("test rank: " + factory.getLowestRank());

        tester.classify(factory, parser, useSeed, min_bootstrap_words, hideTaxon);

    }
View Full Code Here
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
     * @param rank
     * @return
     * @throws IOException
     */
    public static Set<String> randomSelectTaxon(File tax_file, File source_file, float fraction, String rank) throws IOException{
        TreeFactory factory = new TreeFactory(new FileReader(tax_file));
        LineageSequenceParser parser = new LineageSequenceParser(source_file);
        HashMap<String, HashSet> genusTrainSeqMap = new HashMap<String, HashSet>(); // keep the seqID for each genus

        while ( parser.hasNext() ){
            LineageSequence pSeq = parser.next();
            HierarchyTree genusNode = factory.addSequence( pSeq);
            HashSet<String> genusSeqSet = genusTrainSeqMap.get(genusNode.getName());
            if ( genusSeqSet == null){
                genusSeqSet = new HashSet<String>();
                genusTrainSeqMap.put(genusNode.getName(), genusSeqSet);
            }
            genusSeqSet.add(pSeq.getSeqName());
        }
        parser.close();


        // random select nodes at the give rank level
        ArrayList<HierarchyTree> nodeList = new ArrayList<HierarchyTree>();
        factory.getRoot().getNodeList(rank, nodeList);      

        Set<HierarchyTree> selectedNodes = new HashSet<HierarchyTree>();

        int testCount = (int) (((float) nodeList.size()) * fraction);
        while (selectedNodes.size() < testCount){
View Full Code Here
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
    }
   
   
   
    public void calSabSimilarity(String taxonFile, String trainSeqFile, String testSeqFile) throws IOException{       
        TreeFactory factory = new TreeFactory(new FileReader(taxonFile));
        factory.buildTree();
        // get the lineage of the trainSeqFile 
        LineageSequenceParser trainParser = new LineageSequenceParser(new File(trainSeqFile));
        HashMap<String, List<String>> lineageMap = new HashMap<String, List<String>>();
        while (trainParser.hasNext()) {
            LineageSequence seq = (LineageSequence) trainParser.next();
            lineageMap.put(seq.getSeqName(), seq.getAncestors());
           
         }
        trainParser.close();
        NuclSeqMatch sabCal = new NuclSeqMatch(trainSeqFile);
        LineageSequenceParser parser = new LineageSequenceParser(new File(testSeqFile));

        int count = 0;
        while (parser.hasNext()) {
            LineageSequence seq = (LineageSequence) parser.next();
            HashMap<String,HierarchyTree> queryAncestorNodes = getAncestorNodes(factory.getRoot(), seq.getSeqName(), seq.getAncestors());
           TreeSet<KmerMatchCore.BestMatch> matchResults = sabCal.findAllMatches(seq);
           
            short withinLowestRankSab = -1;
            short diffLowestRankSab = -1; 
            String bestDiffLowestRankMatch = null;
            for (KmerMatchCore.BestMatch match: matchResults){
                if ( match.getBestMatch().getSeqName().equals(seq.getSeqName())) continue;
                short sab = (short)(Math.round(100*match.getSab()));
                HashMap<String,HierarchyTree> matchAncestorNodes = getAncestorNodes(factory.getRoot(), match.getBestMatch().getSeqName(), lineageMap.get(match.getBestMatch().getSeqName()));
                boolean withinTaxon = false;
                for (int i = ranks.size() -1; i >=0; i--){                   
                    HierarchyTree queryTaxon = queryAncestorNodes.get( ranks.get(i));
                    HierarchyTree matchTaxon = matchAncestorNodes.get( ranks.get(i));
                    if ( queryTaxon != null && matchTaxon != null){
View Full Code Here
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
        parser.close();
   
    }
   
    public void calPairwiseSimilaritye(String taxonFile, String trainSeqFile, String testSeqFile) throws IOException, OverlapCheckFailedException{       
        TreeFactory factory = new TreeFactory(new FileReader(taxonFile));
        factory.buildTree();
        // get the lineage of the trainSeqFile 
        LineageSequenceParser trainParser = new LineageSequenceParser(new File(trainSeqFile));
        ArrayList<LineageSequence> trainSeqList = new ArrayList<LineageSequence>();
        while (trainParser.hasNext()) {
            LineageSequence seq = (LineageSequence) trainParser.next();
            trainSeqList.add(seq);
         }
        trainParser.close();
        LineageSequenceParser parser = new LineageSequenceParser(new File(testSeqFile));

        while (parser.hasNext()) {
            LineageSequence seq = (LineageSequence) parser.next();
            HashMap<String,HierarchyTree> queryAncestorNodes = getAncestorNodes(factory.getRoot(), seq.getSeqName(), seq.getAncestors());
           
            for (LineageSequence trainSeq: trainSeqList){
                if ( trainSeq.getSeqName().equals(seq.getSeqName())) continue;
               
                HashMap<String,HierarchyTree> matchAncestorNodes = getAncestorNodes(factory.getRoot(), trainSeq.getSeqName(), trainSeq.getAncestors());
                boolean withinTaxon = false;
                String lowestCommonRank = null;
                for (int i = ranks.size() -1; i >=0; i--){                   
                    HierarchyTree queryTaxon = queryAncestorNodes.get( ranks.get(i));
                    HierarchyTree matchTaxon = matchAncestorNodes.get( ranks.get(i));
View Full Code Here
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
            selectedTestSeqIDs = ResampleSeqFile.randomSelectSeq(source_file, fraction);
        }else {
            selectedTestSeqIDs = RdmSelectTaxon.randomSelectTaxon(tax_file, source_file, fraction, rdmSelectedRank);
        }
       
        TreeFactory factory = setup(tax_file, source_file, selectedTestSeqIDs );
        DecisionMaker dm = new DecisionMaker(factory);
         // get all the genus node list
        HashMap<String, HierarchyTree> genusNodeMap = new HashMap<String, HierarchyTree>();
        factory.getRoot().getNodeMap(Taxonomy.GENUS, genusNodeMap);
        if (genusNodeMap.isEmpty()) {
          throw new IllegalArgumentException("\nThere is no node in GENUS level!");
        }
       
        HashMap<String,HashSet> rankNodeMap = new HashMap<String,HashSet>();
        for (String rank: factory.getRankSet()){
            ArrayList<HierarchyTree> nodeList = new ArrayList<HierarchyTree>();
            factory.getRoot().getNodeList(rank, nodeList);
            HashSet<String> nodeNameSet = getnodeNameSet(nodeList);
            rankNodeMap.put(rank, nodeNameSet);
        }

        ArrayList<HashMap> statusCountList = new ArrayList<HashMap>();
        // initialize a list of statusCount, one for each bootstrap from 0 to 100
        for ( int b = 0; b <= 100; b++){
             HashMap<String, StatusCount> statusCountMap = new HashMap<String, StatusCount>();
             statusCountList.add(statusCountMap);
             for (String rank: factory.getRankSet()){
                 statusCountMap.put(rank, new StatusCount());
             }
         }

        int totalTest = 0;
View Full Code Here
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
     * @return
     * @throws IOException
     */
    private TreeFactory setup(File tax_file, File source_file, Set<String> selectedTestSeqIDs) throws IOException {

        TreeFactory factory = new TreeFactory(new FileReader(tax_file));
        LineageSequenceParser parser = new LineageSequenceParser(source_file);

        while ( parser.hasNext() ){
            LineageSequence pSeq = parser.next();
            if ( !selectedTestSeqIDs.contains(pSeq.getSeqName())){
              factory.addSequence( pSeq);
            }
        }
        parser.close();

        //after all the training set is being parsed, calculate the prior probability for all the words.
        factory.calculateWordPrior();
        return factory;
      }
View Full Code Here
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
        if ( index != -1){
            trainsetName = trainsetName.substring(0, index);
        }
        // need to use ISO encoding for UNITE
        FileReader tax  = new FileReader(new File(taxFile));
        TreeFactory factory = new TreeFactory(tax);
              
        LineageSequenceParser parser = new LineageSequenceParser(new File(seqFile));
        LineageSequence seq;
        HashMap<String, String> seqMap = new HashMap<String, String>(); // seqID, desc
        while ( parser.hasNext()){
            seq = parser.next();
            factory.addSequence(seq, false); // donot check the kmers
           
            if ( seq.getSeqName().contains("|S00") ){ // rdpID
                String[] values = seq.getSeqName().split("\\|");
                seqMap.put(values[0], seq.getDesc());  
            }else if (seq.getSeqName().contains("|SH") ){  // if it's seq from UNITE, we need to do something with the seqID            
                String[] values = seq.getSeqName().split("\\|");
                seqMap.put(values[1], seq.getDesc());
            }else {
                seqMap.put(seq.getSeqName(), seq.getDesc());
            }
           
        }
        parser.close();
        HierarchyTreeExtend retVal = new HierarchyTreeExtend(factory.getRoot(), trainsetName);
        retVal.seqMap = seqMap;
        return retVal;
       
    }
View Full Code Here
63
64
65
66
67
68
69
70
71
72
73
   
   
    public PairwiseSeqDistance(String trainseqFile, String taxFile, AlignmentMode mode, boolean show_alignment ) throws IOException, OverlapCheckFailedException{
        this.mode = mode;
        this.show_alignment = show_alignment;
        factory = new TreeFactory(new FileReader(taxFile));
        LineageSequenceParser parser = new LineageSequenceParser(new File(trainseqFile));
       
        while (parser.hasNext()) {
            LineageSequence tmp = (LineageSequence) parser.next();
            seqList.add(tmp);
View Full Code Here
TOP

Related Classes of edu.msu.cme.rdp.classifier.train.validation.TreeFactory

  • edu.msu.cme.rdp.classifier.train.validation.crossvalidate.CrossValidate
  • edu.msu.cme.rdp.classifier.train.validation.crossvalidate.RdmSelectTaxon
  • edu.msu.cme.rdp.classifier.train.validation.distance.CompareTrainingSets
  • edu.msu.cme.rdp.classifier.train.validation.distance.PairwiseSeqDistance
  • edu.msu.cme.rdp.classifier.train.validation.distance.TaxaSimilarityMain
  • edu.msu.cme.rdp.classifier.train.validation.leaveoneout.LeaveOneOutTesterMain
  • edu.msu.cme.rdp.classifier.train.validation.movingwindow.MainMovingWindow
  • java.io.BufferedReader
  • java.util.StringTokenizer
  • java.util.List
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.